AIPS HELP file for BPSMO in 31DEC18
As of Mon Apr 23 23:11:15 2018
BPSMO: Smooths/interpolates bandpass tables to regular times
INNAME Input UV file name (name)
INCLASS Input UV file name (class)
INSEQ 0.0 9999.0 Input UV file name (seq. #)
INDISK 0.0 9.0 Input UV file disk unit #
BPVER Bandpass table version
BCHAN -1.0 8192.0 Lowest normalization channel
-1 = none, 0 = center 3/4
ECHAN 0.0 8192.0 Highest normalization channel
DOWEIGHT -2.0 2.0 Use weights in smoothing?
APARM Control information:
(1) function type
(2) function diameter (hours)
(3) function support (hours)
(4) Output interval (hours)
(5) >0 = vector normalization
Use: To create a new 'BP' extension file containing a bandpass
correction solution for each antenna at each of a set of
regularly spaced times or at the input times but with flagged
solutions replaced by time-smoothed ones. A variety of
interpolation and smoothing options are provided. The amplitude
of the solutions may be forced to one over a range of channels
with either scalar or vector averaging (or it may be allowed to
INNAME.....Input UV file name (name). Standard defaults.
INCLASS....Input UV file name (class). Standard defaults.
INSEQ......Input UV file name (seq. #). 0 => highest.
INDISK.....Disk drive # of input UV file. 0 => any.
BPVER......Specifies the version of the BP table to be read as input.
0 -> highest. The output version is always highest + 1.
BCHAN......Lowest channel of a range of channels used to determine
the amplitude normalization. 0 => bchan = nchan/8 and
echan = (7/8)*nchan (user ECHAN ignored). < 0 => do no
ECHAN......Highest channel of a range of channels used to determine
the amplitude normalization. NO DEFAULT when BCHAN > 0.
DOWEIGHT...> 0 => use the solution weights in the time smoothing; else
ignore them. If the solution weights are very variable,
but the solutions have real variations with time, then it
is best to ignore the weights. If the weights are not
greatly different, then using them improves the noise of the
smoothed result. DOWEIGHT = 2 or -2 causes the output
weights to be computed by noise combinatorial methods,
otherwise the data weights are interpolated. If the
solutions do not vary rapidly with time, then an averaged
solution is more reliable and DOWEIGHT 2 would reflect
that. If they do vary with time, then the nearest neighbor
gets less reliable (for estimating the BP at the present
time) the further away it is in time. DOWEIGHT -1 (or 1)
is a partial compensation for this.
APARM......Specifies the type of time smoothing to be applied to the
APARM(1) = type of smoothing to apply:
0 => Two-point (linear between nearest neighbors)
1 => Hanning (linear)
2 => Gaussian
3 => Boxcar
4 => Sinc (i.e. sin(x)/x)
APARM(2) = the "diameter" of the function in hours, i.e.
width between first nulls of Hanning triangle and sinc
function, FWHM of Gaussian, width of Boxcar
0 => 1 hour.
APARM(3) = the diameter over which the smoothing function
has value in hours. Defaults: 1, 3, 1, 4 times APARM(2)
used when input APARM(3) < net APARM(2).
APARM(4) = increment between output bandpass solutions in
hours. 0 -> 1 hour.
< 0 => the output BP file has the same times as the
input and has the same bandpasses, except that IFs or
polarizations that were flagged are replaced with ones
computed by smoothing using the other APARM parameters.
APARM(5) = flag for type of normalization and smoothing.
<= 0 => scalar averaging of amplitudes,
> 0 => vector averaging.
In cases of decent S/N, use scalar.